# Questions tagged [sensitivity-analysis]

Auxiliary methods intended to check if the outcome of an analysis strongly depends on the model assumptions, preprocessing steps, presence of outliers, etc.

164 questions
Filter by
Sorted by
Tagged with
31 views

### How would I do sensitivity analysis for other models? (logistic regression, random forest, Cox proportional hazard) [closed]

I wanted to know how to conduct sensitivity analysis for causal inference on my models. Right now I've used logistic regression, a random forest model, and in another study I have used a Cox ...
1 vote
17 views

### Correlated parameters

If I use a regression model for estimating a time-to-event distribution function with several parameters (such as a Weibull), should I consider that those parameters are correlated when I later ...
• 43
26 views

### Rank the sensitivity among multiple calibration curves

Let's say that, with a measurement device, we have a linear relationship between an output measurement I (in mV) and the concentration ...
• 632
19 views

### How can I provide meaningful commentary about the uncertainty associated with a population estimate drawn form individual ML predictions?

Context: Suppose a team develops a prediction model that predicts the presence of a condition for a given individual. This model is trained and externally validated before being picked up by a ...
1 vote
56 views

### Sensitivity analysis of features in a Random Forest Model

I have a built a large Random Forest Classifier and was able to output the feature importance as below: I understand that this importance is a based on mean decreased impurity. But how to interpret ...
• 11
46 views

### Compute Sobol index from scratch

0 I am trying to compute Sobol index by myself using python but I am facing some problems. The formula is very simple however I don't understand some part : Here, E[Y|Qi] denotes the expected value ...
• 1
30 views

### Global sensitivity indices with sign (positive/negative)

I’ve used Sobol global sensitivity indices to analyze the effect that some input variables have on a nonlinear regression model, but now I need to know if the effect is positive or negative (i.e. if ...
• 497
1 vote
83 views

### Sensitivity Analysis for Raw Model Output or Model Predictions?

For a variety of reasons, many researchers have suggested that, for attempting to make causal inferences with non-linear statistical models, one should generally avoid endowing a causal interpretation ...
32 views

### Nonlinearity of model using Sobol indices

I'm analyzing a computationally demanding numeric model where I want to show that nonlinearities play a certain role for my problem. I want to do this using Sobol sensitivity indices of first oder by ...
• 41
39 views

### Multiple imputations generate values distributed differently from original dataset... does this mean my data is MNAR? Imputations still usable?

Quick question. I'm using the mice R package to impute missing data. I go by the presumption that the missing data are MAR, but I wouldn't be surprised if a few binary variables were MNAR. I followed ...
• 61
23 views

### Survivability Analysis whereby study drop put/ censoring indicates a positive result

I am trying to conduct a survivability analysis (particularly a Kaplain-Meier curve) on data for estimating the likelihood of university drop-outs over time. The study began in 2015 (accounting for ...
34 views

### Best way to address selection bias when outcome cannot be randomized

I have an (low incidence) binary outcome compared between 2 groups. The intervention for group 1 is coming from a specific type of center (academic) while group 2 from a different center. It is not ...
17 views

### Dataset to apply Sobol Indices

I am working on a project based on Sobol indices. I am looking for a dataset to apply Sobol indices. I need to show a basic demonstration of this method. Specifically, I want to identify which inputs ...
• 175
390 views

### How to interpret a sensitivity analysis from G*Power for a repeated measures, within factors ANOVA

I am trying to conduct a sensitivity power analysis for a (2x2) ANOVA. The participants all received trials based on color or shape, and items could be either old or new. In G*Power I have the ...
• 153
18 views

### Sensitivity Analysis selection for nested data?

So the data I'm working with is GPS tracking data of pelicans and Offshore rigs with the goal of seeing if pelicans are positively or negatively affected to the presence of rigs. But before I can do ...
55 views

### What are the current guidelines for performing sensitivity analysis on matched data that is able to compare bias in reference to other covariates?

I am trying to perform sensitivity analysis on a causal inference (observational study) problem in a dataset that has a binary outcome and binary treatment. I've applied matching and g-computation ...
• 133
22 views

### Test if difference between two groups across 4 categories [duplicate]

I want to make a sensitivity analysis to test whether there is a statistically significant difference between the observations included for the study (n = 100,000) and the excluded observations (n = ...
• 139
14 views

### Interpretation of sensitivity analysis provided by tgp package

I have a question about the interpretation of the sensitivity analysis I obtained by R's tgp package (https://cran.r-project.org/web/packages/tgp/vignettes/tgp2.pdf) as can be seen below. The relation ...
• 1
51 views

### How to deal with missing value in dependent variable of prediction model?

I am trying to build a prediction model from longitudinal study after intervention. So after intervention, we followup patients 1,3, and 6 months later to see if they are cured or not. So dependent ...
42 views

### Two-sample Kolmogorov Smirnov test for global sensitivity analysis - How to treat discrete distributions?

Dear Cross Validated community, We are working on a uncertainty & sensitivity analysis using a mathematical optimization model. More specifically, we have a set of uncertain parameters, which ...
• 101
13 views

### Sensitivity analysis of a fuzzy cognitive model

I have a fuzzy cognitive model of inter-organizational collaboration that is represented by a 27x27 matrix. I want to analyze the effects of individual variables. I've read I can do this best with a ...
1 vote
33 views

### Comparing Sensitivities of Microscopic and Macroscopic Models (Travel Demand)

A simulation output for each microscopic travel demand model and two further macroscopic MDCEV models - transformed versions of the microscopic model - are given for a base scenario. Moreover, three ...
• 31
36 views

### Is Sensitivity Analysis for Making Causal Inferences Only for Backdoor Adjustment?

I am wondering if sensitivity analysis for causal inference is only applicable when doing backdoor adjustment/selecting on observables. Conventionally, sensitivity analysis evaluates the threat of an ...
100 views

### Distribution of Sobol's Indices

Some background: Given a linear regression model (or any other GLM), we all know how to test the null hypothesis $\hat\beta_i=0$. The lm function in ...
• 4,032
1 vote
192 views

### Bounds of Shapley values for variable importance

Imagine you have either a very good predictive model $f$ for a response $y$ or two highly predictive models $f_1$ and $f_2$. Is it possible to bound the "true" Shapley values of $y$ in ...
• 2,708
90 views

### Generalizability of Sensitivity Analysis Results

I have executed an analysis using the novel approach developed by Imai et al. (2021) that allows for the execution of matching/weighting techniques when using panel data. I was planning on running ...
1 vote
26 views

### Sensitivity analysis based on a dataset with dependent variables

I am part of a company that produces data from satellite images and machine learning/deep learning. We produce different data and sometimes the results of one step will be used as input for the next ...
1 vote
383 views

### Updated: How to do a sensitivity analysis for an ordered logistic regression in R?

How would you do an robustness or sensitivity analysis for an ordered logistic regression? Can it be done by replacing the control variables with others that are similar in the model? For instance ...
• 65
144 views

### How can I learn to refutation methods in DoWhy Library?

The refutation methods in the dowhy causal inference library are useful. https://www.pywhy.org/dowhy/v0.8/dowhy.causal_refuters.html I'd like to learn a theory that explains these methods well, but I'...
• 73
1 vote
27 views

### Does Rubin-Rosenbaum sensitivity analysis include selection bias as special case?

According to pg 2 of the pdf http://www-stat.wharton.upenn.edu/~rosenbap/BehStatSen.pdf, "however, this alternative method takes account of sampling variability and is applicable to any kind of ...
• 1,445
460 views

### What is the difference between the sensitivity analysis and feature importance

As I understand in the sensitivity analysis, the variables are perturbed and their effect on the output is observed. In feature importance, we want to find the same answer, but with different methods ...
63 views

### How do I complete a sensitivity analysis to simulate censored data?

I am currently trying to analyze the duration of the egg stage of two species of insect (factor 1; 2 levels - HA and AP) at several different temperatures (factor 2: 5 levels - 20, 23, 26, 29, 32) ...
23 views

### Sensitivity to settings vs statistical sensitivity

I am confused between two different properties of a model. Both are called sensitivity. First, sensitivity is another word for true positive rate. Second, the output of a model changes in response to ...
• 97
363 views

### Log of a log-transformed variable

I have been suggested to use the log of a log-transformed independent variable (i.e., log(log healthcare expenditure)). I am not sure how would this make sense. Is this a standard practice (in the ...
• 71
54 views

### Does it make sense to do a prior sensitivity analysis if using flat priors (Mplus default)?

If it does make sense to do a sensitivity analysis how should one determine which priors to use? If flat "non-informative" priors are chosen to begin with because of a lack of information ...
• 11
1 vote
96 views

### R2 limitation for linearity and monotonic?

I'm studying sensitivity analysis. I know that to use PCC and SRC, linearity and/or monotonic must be assumed. So I'm trying to calculate R² for this, and my question for this is: Can I use GLM to ...
• 35
1 vote
29 views

### A few queries regarding Meta-analysis and its subparts using R

As part of a meta-analysis of a disease in R, I received some feedback on the paper of which it is a part by external consultants. I cannot contact them again, and am confused about the advice they ...
26 views

### Understanding "In Bayesian inference, the difference between data and a parameter is that one is observed (data) and one isn't (parameter)" [duplicate]

In his statistical rethinking course, Richard Mclreath states "In Bayesian inference, the difference between data and a parameter is that one is observed (data) and one isn't (parameter)" I ...
• 584
62 views

### Inferring effect & effect modification from simulation data

I have a "black box" system (computer simulation), which takes inputs: $x_1 \in [0,1]$, $x_2 \in [0,1]$, and $N_i$ others $\vec\theta = \{\theta_1, \dots, \theta_{N_i}\}$, and produces an ...
• 215
155 views

### When do we do sensitivity analysis in biostat and how do we do it?

I have two questions below. I have seen people doing sensitivity analysis in observational study papers for the model to check sensitivity to assumptions in bayesian context for selection of priors. I ...
• 1,445
1 vote
26 views

### Can a global sensitivity analysis be performed on Bayesian inference?

My question is, is it possible to perform a Global Sensitivity Analysis on a Bayesian inference model (not just on the prior, the entire model)? A bit of context: I am fairly new to Bayesian ...
347 views

### How to conduct sensitivity analysis on IPSW for survival data?

I am working on survival data, comparing two groups of patients (with or without treatment). These data have some selection bias, thus, I have chosen to weight my sample by Inverse Propensity Score ...
1 vote
181 views

### Is there a method for Global Sensitivity Analysis that does not require special sampling methods?

I want to perform a global sensitivity analysis using randomly sampled data that already exists (or can be generated with only N randomized model runs). The impetus for this is to be able to use the ...
• 173
234 views

### Interpretation of the E-value for non-epidemiologists

A recent method for sensitivity analysis is the E-value (VanderWeele and Ping, 2017). Yet, I'm still struggling with the interpretation of such a value. Coming from outside of epidemiology, where risk ...
• 269
200 views

### Trade-off between omitting variables or dropping observations in multivariate logistic regression

Say you are selecting $n$ observations from a complex survey of $N$ individuals to create an analytical sample of relevant observations; and that you intend to fit a binomial multivariate logistic ...
• 23
1 vote
45 views

### What-If Scenario Regression Modelling

I'm pondering a scenario involving some insurance data but this could be relevant in many fields. The idea is that I have a total count of some event. Let's imagine this count is the # of attorney ...
• 181
1 vote
86 views

### what is the sensitivity of the neural network using standardized input

Suppose I trained a neural network with standardisation of the data following (X-EX)/std(X). The input is x(t) and output is y(t). How can I calculate the sensitivity of this trained network (...
• 163
32 views

### Can I perform sensitivity analysis, if I don't know expected prediction results?

Can I perform sensitivity analysis, if I don't know expected prediction results? I.e. I have a model with input parameters and weights. But I don't know when a prediction should be true and when false....
• 4,119
2k views

### How do you deal with A/B testing for small samples?

I am performing A/B testing (basically hypothesis testing) with relatively small samples, so the results are largely inconclusive. I am aware of techniques like CUPED (for decreasing the sample ...
• 23